Online auctions and the winner's curse.

Online auctions are the Internet flavor of the moment. The best-known auction site, eBay, went public in September 1998 at $18 and shot up to $47 by the end of its first day. It's now at about $150, which means that investors value the company at around $20 billion despite only $47 million in sales over the last 12 months. Traditional businesses want in on online auctions as well. You have until March 21, 1999, for instance, to bid on the pink satin eye mask Faye Dunaway wore in Mommie Dearest (current high bid: $650) at Universalstudios.com.

A technology that allows you to bid on a 1970 Château Lafite-Rothschild (Wine.com, $205) without leaving your study is impressive. But it cannot repeal the laws of economics. One such law is dubbed the "winner's curse" and holds that the winner of an auction almost always overpays. As an understanding of this law makes clear, online auctions make the winner's curse even worse.

Auctions are often thought of as models of economic efficiency, uniting buyers and sellers at just the right price to maximize their mutual satisfaction, put resources to their highest and best use, and so on. But three petroleum engineers writing in the Journal of Petroleum Technology explained in 1971 why this is not the case.

Suppose several petroleum firms are bidding on the drilling rights to a piece of tundra. No firm is sure how much oil is underneath the property, so they hire a team of engineers to poke at the surface rocks and make a guess. The guesses will likely range from too low to too high. Some firm's engineer will probably guess right, but that firm won't win the auction. The winner will be the firm whose engineer was the most overoptimistic. The winning firm won't ultimately get as much oil as their engineers promised, meaning the firm paid too much. In short, the auction "winner" is ultimately a loser.

This is a particularly clear example because the thing being auctioned will have a definite value in the future that is unknowable at present. But the winner's curse afflicts auction bidders whenever there is uncertainty over the current or eventual value of the item on the block. This is true even when bidders have no intention of reselling the item and when its innate value seems inherently subjective.

For example, bidders for Faye Dunaway's pink eye mask must make some judgment on how much they care about Faye Dunaway. If that were all, the winner would likely be the person who cared the most. That would be economically efficient in two senses: 1) the utility of Dunaway's eye mask would be maximized by placing it with the person who can extract the greatest pleasure from it (just as, uncertainty aside, the highest bidder for an oil field will be the person who can extract the most oil from it); and 2) that person would pay no more for the eye mask than the pleasure of owning it was worth to him.

But the course of love is as uncertain as the petroleum content of a pile of rocks. Bidders must also try to guess how much they'll care for Faye Dunaway in, say, 10 years. The more you overestimate your undying affection, the more likely you are to win the auction--and the more likely you are to feel like an idiot in 2009.

Economists have pointed out that if bidders were truly rational, they'd simply reduce their bids to correct for the winner's curse. There is even a mathematical proof that a perfectly rational actor can avoid the curse. But experimental evidence suggests that even experienced bidders don't reduce their bids by enough. For instance, a study of oil field auctions shows that even seasoned firms typically pay far too much for drilling rights given the amount of oil they eventually recover. The same phenomenon has been observed when corporate takeover wizards bid on other companies--the "winner" often overpays. In other words, oil firms and corporate takeover specialists keep on getting burned in auctions but persist in bidding too high. They simply don't learn.

I rritatingly, a rational person who understands the winner's curse can't do anything about it so long as the other bidders continue to bid irrationally. If you bid rationally (lower), you won't win any auctions; if you bid what it takes to win auctions (higher), you'll lose money because of the winner's curse. Economist Richard Thaler wickedly suggests a solution: Explain the theory to your competitors. He posits that this is exactly why the three oil engineers published their article explaining the curse in 1971. Their hope was to induce other firms to reduce their bids. If so, it didn't work, since oil firms continue to overpay.

Online auctions worsen the winner's curse by increasing the number of bidders. The craziest poor sucker in a group of 20,000 bidders on the Internet is likely to be crazier than the craziest one among 200 in a Burbank hotel ballroom. That's another thing that experimental economists have confirmed--the larger the group, the bigger the winner's curse. There's no satisfactory way to buy rare or one of a kind items, but online auctions are a particularly bad method.